Abstract: In order to account for the effect of various factors on acid orange 7(AO7)decoloration in a pulsed discharge plasma(PDP)system with a multi-needle-to-plate electrode,the BP neural network model is used in this paper to simulate the oxidation process of PDP under different reaction conditions. This review focuses on the changing trends of AO7 decoloration efficiency under different solution conditions,including initial concentration,initial electrical conductivity and initial pH of AO7 solution.The results show that the differentials between the predicted values obtained from the established BP neural network model and the experimental values are within ?5%,so the BP neural network model can be used to forecast the influence rule of solution conditions on decoloration rate.The analytic results indicate that AO7 decoloration rate is higher under certain initial solution concentration,solution conductivity has little effect on AO7 decoloration efficiency and acid solution condition is favorable for AO7 degradation in the PDP system.

Key words: pulsed discharge plasma, acid orange 7, decoloration, BP neural network, model

摘要： 为了说明多针-板电极脉冲放电等离子体(PDP)氧化酸性橙Ⅱ(AO7)过程中多种因素对AO7脱色效果的影响,本研究建立了PDP氧化AO7的BP神经网络模型,用BP神经网络模拟多因素作用下PDP氧化AO7的反应过程,重点考察溶液条件,包括溶液初始浓度、初始电导率和初始pH等条件对AO7脱色效果的影响趋势。研究所建立的BP神经网络模型对AO7脱色效率的预测值与实际测定值的偏差在±5%范围内,说明BP神经网络模型可以很好地分析PDP这一复杂体系中各溶液条件对AO7脱色效率影响的规律。模型分析结果表明:所建立的PDP水处理系统中,在一定的AO7溶液初始浓度条件下有较高的脱色效果;溶液电导率对AO7的脱色效果基本没有影响;酸性条件更有利于AO7的脱色。

关键词: 脉冲放电等离子体, 酸性橙Ⅱ, 脱色, BP神经网络, 模型